Three observations prompted recent surge of interest in bacteriophage (phage) biology: (1) many virulence factors in pathogenic bacteria are carried by phages, (2) there is a high rate of horizontal gene transfer among phages, and (3) there are a lot of phages in the biosphere. The corollary of these observations is that there is a large reservoir for these virulence factors and even phylogenetically remote phages can potentially get access to them. Though we have detailed knowledge on phages, we generally have little understanding on how ecological and evolutionary factors influence the fate of a phage. Instead of treating the phage fitness as a complex trait, this proposed research program integrates detailed knowledge on phages to decompose phage fitness into various interacting parts of phage life history traits. The effects of ecological factors and genetic environments on phage fitness can be tested against a priori models. The long-term goal of this proposed research program is to understand phage genome evolution through life history traits evolution. A multi-disciplined approach is proposed to study one of the traits: the timing of bacterial host lysis, and how such a decision will impact the phage's fitness. Three Specific Aims are proposed: (1) Competition experiments between isogenic phage strains will be conducted to demonstrate the importance of lysis timing under different ecological conditions and the context of other life history traits. Long-term evolution experiments will be conducted to show that lysis timing is an evolutionarily malleable trait. Multiple infection of phage strains will be conducted to assess the impact of intra-host competition on the evolution of lysis timing. (2) Wild phages will be isolated from Australia and Mexico to determine genetic variations in lysis genes, lysis times, and fitness. Within- and between-population comparisons will be used to test the hypothesis that lysis timing is also important in the wild. (3) The evolutionary dynamics of holing protein sequence is tackled using in vitro constructed lambda library. The lysis time distribution will be determined for a given sequence and its variants 2- or 3-mutational steps away. Highly diverged but functional holing sequence will be evolved. Competition of the lambda library will be conducted to select for the fittest genotype. Possible evolutionary passages between two locally adapted genotypes will be reconstructed.